Graph Chat

Introducing our new Graph Chat feature, designed to streamline how you explore and question your data by chatting directly with your current subgraph. This feature leverages advanced language models to combine your list of genes and focus topics with the live structure of your knowledge graph, delivering evidence-backed answers grounded entirely in your network. You can ask short or complex questions, and depending on the Resource you’re accessing (Free or Premium), you have a monthly limit on the number of chat interactions you can run (again, depending on your plan).

Back to nova
new
From Search to Conversation: Graph Chat
Our new Graph Chat feature transforms how you interact with the knowledge inside your current subgraph. By leveraging a specialised version of Anthropic Claude, it enables real-time, conversational exploration of the genes, traits, phenotypes, publications, and links in your graph. Instead of static summaries, you can now ask dynamic questions and refine your understanding interactively.
Fine Tuned & Enhanced
We’ve put significant effort into reducing hallucinations by fine-tuning our prompts and using a strict Retrieval-Augmented Generation (RAG) approach with low temperature. Graph Chat only retrieves and reasons over content found within your current subgraph. This allows you to add or remove nodes and edges to modify the context the model is basing it's answers on. The model is instructed to rely exclusively on this retrieved graph evidence when generating answers, ensuring responses remain fully grounded in your data.

Real-World Impact

Graph Chat has already proven valuable in real-world research workflows, with users such as NRC Canada highlighting the impact of being able to hold natural, conversational interactions directly with their knowledge graph. Instead of manually traversing dense and complex networks, researchers can ask precise follow-up questions, refine their queries in real time, and rapidly drill down into supporting evidence. This conversational layer enables users to extract deeper biological detail faster, uncover indirect and non-obvious relationships, and explore alternative hypotheses with far less friction. As a result, teams are moving more efficiently from data interrogation to interpretation, and from insight to confident, defensible decisions.

Read more on our News Article

We’ve worked on these updates alongside the experts who use them most and believe they make KnetMiner a more powerful, user-friendly tool, tailored for those who need results fast.
‍
‍We welcome any and all positive feedback as we begin rolling out the all New KnetMiner.

Explore NovaContact Us